<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
        <title>Affaire Thélyson Orélien : les détecteurs d’IA sont-ils fiables ?</title>
        <link>https://vid.freedif.org/videos/watch/1b77afea-5f9f-4f72-bba2-9496c31f65e7</link>
        <description>La polémique autour du livre de Thélyson Orélien a relancé l’intérêt pour les logiciels capables de détecter les textes générés par l’IA. Mais ces outils, qui analysent notamment la structure des phrases, le vocabulaire et la prévisibilité des mots, sont loin d’être fiables à 100 %. Le risque de faux positifs est réel, notamment lorsque les textes ont été retravaillés par un humain ou présentent un style d’écriture particulier.</description>
        <lastBuildDate>Tue, 29 Sep 2026 06:35:50 GMT</lastBuildDate>
        <docs>https://validator.w3.org/feed/docs/rss2.html</docs>
        <generator>PeerTube - https://vid.freedif.org</generator>
        <image>
            <title>Affaire Thélyson Orélien : les détecteurs d’IA sont-ils fiables ?</title>
            <url>https://vid.freedif.org/client/assets/images/icons/icon-1500x1500.png</url>
            <link>https://vid.freedif.org/videos/watch/1b77afea-5f9f-4f72-bba2-9496c31f65e7</link>
        </image>
        <copyright>All rights reserved, unless otherwise specified in the terms specified at https://vid.freedif.org/about and potential licenses granted by each content's rightholder.</copyright>
        <atom:link href="https://vid.freedif.org/feeds/video-comments.xml?videoId=1b77afea-5f9f-4f72-bba2-9496c31f65e7" rel="self" type="application/rss+xml"/>
    </channel>
</rss>